Alerts That Save Crops: Configuring MultiGrow Alarms and Diagnostics
Alerts That Save Crops: Configuring MultiGrow Alarms and Diagnostics
Most growers don’t dream of dashboards. They want a nudge before things go sideways—and a short, obvious path to a fix. That’s what good alarms and good graphs do: turn a ten-hour mystery into ten minutes with a phone and a wrench. MultiGrow’s control logic runs locally—even if your 4G/5G hiccups, the house keeps obeying the plan. The cloud is there to watch, analyze, and nudge. The trick is deciding which nudges matter.
Fewer, Smarter, Actionable
An alarm is not a reminder. It’s a request for action. Start there and you’ll configure less noise and get more done. For most houses the “starter set” is simple: wind and rain (to protect vents and scheduling), inside temperature and humidity (so you can manage VPD and avoid disease-friendly conditions), EC/pH where you’re mixing nutrients, and a few plumbing basics like tank levels or flow so you catch “we’re dosing air” before plants tell you.
This aligns with a point your engineers made repeatedly: some customers turn off alerts because they’re noisy, then call support when the inevitable happens. Configure fewer, smarter alarms and they’ll stay enabled—and useful.
Thresholds that Behave Like Your Structure
The same wind gust that’s harmless on a low house can be nasty on a tall one. Set conservative wind thresholds that reflect your structure and covering, then test. If you saw vents slam shut on a sunny afternoon during a brief squall, add a short delay or tweak the trigger so you’re not “chattering” hardware because of noise. The goal isn’t to eliminate protective behavior; it’s to make sure it’s proportional.
Apply the same thinking to EC/pH. If your tolerance is tight, make sure alert bands are also tight. If you’re comfortable with wider drift during a specific stage, don’t leave alarms at a factory default that will ping you for behavior you’ve already accepted.
Three Diagnostics You’ll Actually Use
Alarms are the front door. Diagnostics is the living room. When something trips, graphs make the next step obvious.
1) Vents closed on a warm day, house overheated.
Plot wind speed and vent position for the window around the event. If you see a spike followed by vent closure, the system did exactly what you asked. The fix is usually a threshold tweak or a small delay, not a parts run. This example came straight from support: once the grower could see wind and vent movement together, the “mystery” evaporated.
2) EC won’t climb, even though dosing is “on.”
Overlay EC with dosing output state and stock tank level. When the log shows the controller commanding output and the graph shows no EC response, it’s almost always physical: an empty tank, a clogged injector, or a manual valve left shut after refill. Your team has seen all three, many times. In one operation, the same closed valve triggered identical pH alarms every other Friday—until someone followed the data to the lever.
3) Night humidity creeps high, disease risk climbs.
Plot temperature, humidity (or VPD), and fan states across the night. If you irrigate late, you may be adding moisture faster than you can remove it. The fix can be as simple as skipping a late cycle in those conditions and increasing airflow overnight. In coastal or monsoon regions, the system can actively reduce or skip irrigation on humid nights and ramp circulation to keep conditions out of the Botrytis comfort zone.
Notifications That Reach Humans
Delivery matters. For critical alarms, use SMS and email. Escalate if nothing is acknowledged after a short window. Test the path monthly—simulate a wind event and make sure the right phones buzz. Keep the contact list current. It sounds obvious until the person on vacation is the only one getting the message that your rain plate thinks the sun is shining.
And remember: because MultiGrow’s control logic runs locally, connectivity blips don’t stop the greenhouse from running. Alarms and graphs may be delayed; plants won’t wait for the internet.
Clean Data or Don’t Bother
We’ll keep banging this drum: alarm quality follows sensor quality. If the weather station isn’t maintained, a “no rain” assumption will happily run irrigation through a drizzle. If probes drift, EC alarms become coin flips. The maintenance tasks aren’t glamorous—clean the rain plate monthly, keep optics clear, calibrate EC/pH on a sane cadence—but they’re the difference between trust and fatigue.
Train for Calm, Not Heroics
A 30-minute refresher each quarter pays all year. Review which alarms you actually acted on last season. Tighten the ones that mattered; retire the ones that never did. Make sure the right people get the right notifications. Practice one manual override so that in a true edge case, nobody is learning while adrenaline is high. Your engineers mentioned a 500-page manual for climate control; nobody wants to read that in a storm. A short ritual beats a long binder.
The Point of the Exercise
Good alarms don’t make you busier. They make you calmer. They give you a few clear chances to intervene before a drift becomes damage. They help your team tell the difference between “ignore it” and “move now.” And when something does break, the graph tells you where to look first.
If you want a head start, ask for our one-page Alarm Setup Starter and a short remote walkthrough. We’ll tune thresholds for your structure, your crop, and your weather reality, and make sure the messages go to people who will actually move. The greenhouse will keep running either way—that’s what local control is for. But with a few smart nudges, you’ll spend more time growing and less time guessing.